I00
Journal of General Microbiology (1975), 90, I 00-1 14
Printed in Great Britain
The Hydrogenation of Unsaturated Fatty Acids by Five Bacterial
Isolates from the Sheep Rumen, Including a New Species
By P. K E M P , R. W. W H I T E A N D D. J. L A N D E R
Department of Biochemistry, ARC Institute of Animal Physiology,
Babrdham, Cambridge CB2 4AT
(Received 5 February 1975; revised 26 April 1975)
SUMMARY
Five strictly anaerobic bacteria able to hydrogenate unsaturated fatty acids
were isolated from sheep rumen. One was characterized as Ruminococcus albus,
two as Eubacterium spp. and two as Fusocillus spp., one of which is named as a new
species. The Fusocillus organisms were able to hydrogenate oleic acid and linoleic
acid to stearic acid, and linolenic acid to cis-octadec-15-enoic acid. The R. albus
and the two Eubacteria did not hydrogenate oleic acid but converted linoleic and
linolenic acids to a mixture of octadecenoic acids ; trans-octadec- I I-enoic acid
predominated but several isomeric cis and truns octadecenoic acids were produced
together with isomers of non-conjugated octadecadienoic acids.
The intermediate and final products of hydrogenation by each organism were
compatible with the results from mixed rumen bacteria.
INTRODUCTION
The hydrogenation of dietary fatty acids in the rumen has been reviewed in detail by
Dawson & Kemp (1970) and by Viviani (1970). The identities of the major intermediates
in the hydrogenation of a-linolenic acid and linoleic acid were established by in vitro
incubations of ~J~C-labelled
substrates (Wilde & Dawson, I 966 ; Dawson & Kemp, I 970).
Noble, Steele & Moore (1969) showed that the same major pathways operate in vivo.
Efforts to obtain active preparations from the broken cells of mixed populations of rumen
organisms have been unsuccessful, but Kepler & Tove (I 967) tried to obtain active enzyme
systems from disrupted cells of a strain of Butyrivibrio jibrisolvens, a rumen bacterium
which had earlier been shown to hydrogenate both a-linolenic and linoleic acid to transoctadec-1 I-enoic acid (Polan, McNeill & Tove, 1964; Kepler et al. 1966). Only the isomerase which catalysed the first step in the process was isolated. Subsequent work (Kepler,
Tucker & Tove, 1970) demonstrated that a double bond in the 9-10 position and a free
carboxyl group were necessary for hydrogenation, as was found by Garton, Lough &
Vioque (1961) and Hawke & Silcock (1969) who showed that the fatty acids of triglycerides
were hydrogenated only after lipolysis. Recently this has been shown to be true for the fatty
acids of the galactosylglycerides of grass (Dawson et a/. 1974). Attempts to find the source
of reducing power and the immediate hydrogen donor in mixed populations (Viviani, 1970)
and in B.jibrisolvens (Rosenfeld & Tove, 1971) met with little success.
Activity was detected by Mills et al. (1970) in a micrococcus isolated by Russell & Smith
(1968) and by Sachan & Davis (1969) in a Borrelia sp. Both organisms, like B.jibrisoZvens,
produced trans-octadec- I I -enoic acid from a-linolenic and linoleic acids.
In 1968 we began a programme of systematic screening of rumen bacteria for their ability
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 03:04:38
Hydrogenation by rumen bacteria
I01
to hydrogenate linolenic acid, linoleic acid and oleic acid. We have examined more than 200
isolates. Thirty showed limited activity, and of these only five had sufficient activity to
warrant further work.
The isolation in pure culture, the identification of these five bacteria (one as a new
species) and the separation and characterization of their hydrogenation products is
described in this paper.
Some preliminary aspects of this work have been reported (Kemp & White, 1968; White,
Kemp & Dawson, 1970; White & Kemp, 1971).
METHODS
Isolation and testing of bacteria. For initial isolations, subsequent culturing and for all
hydrogenation activity testing, the medium of Bryant & Robinson (1961)with the addition
of 0 . 1 % (w/v) yeast extract was used, the agar concentration being increased to 1-25 %
(w/v) for roll bottles. This medium is referred to throughout as the basic medium. All tubes
and roll bottles of basic medium were gassed with 0,-free CO, and closed with butyl rubber
bungs [Charles W. Meadows (London) Ltd, Torquay, Devon]. Incubations were initially
for 72 h at 39 "C, but since it was found that viable bacterial numbers were maximal at
about 20 h subsequent hydrogenation incubations were not usually carried beyond 24 h.
For colony isolation and for any situations where other reducing mechanisms might not
be operative, or where aeration risks during handling were high, levels of L-cysteine were
increased from the 0.25 mg/ml normally used, to as high as 2 - 5 mg/ml.
Utilization of carbohydrates, alcohols and organic acids was tested by substituting
them, at a concentration of 5 mg/ml, for the glucose, starch and cellobiose normally added
to the basic medium. Unequivocal growth was accepted as a positive result.
Culture fluids with HC1 added to 0 . 1 M, were analysed directly for volatile fatty acids
using a Tween-80 column (20 % on gas-chrom 2; Pye-Unicam Ltd) at 130 "C. Lactic and
succinic acids were estimated as their methyl esters by g.1.c. at 75 and 130 "C respectively,
using a 10% PEG A column. In addition, lactic acid was estimated enzymically by the
method of Barker & Britton (1957).
De-carboxylases, arginine dihydrolase, catalase and the reduction of gluconate and
nitrate were all tested far according to the methods described by Cowan & Steel (1966),
under anaerobic conditions. Casein hydrolysis was tested in roll bottles of basic medium
containing 0.5 % (w/v) light white casein. Liquefaction of gelatin was tested by adding
gelatin-charcoal discs (Oxoid) to tubes of basic medium. Hydrogen sulphide formation was
detected by the method of Bryant & Small (1956), except that the ferrous salt was substituted
for the ferric and it was sterilized under nitrogen.
Rumen fluid, both as inoculum material and as a constituent of the basic medium, was
obtained from Clun Forest sheep, each having a permanent rumen cannula. The animals
were permanently housed, and received a maintenance diet of 1000g hay chaff and IOO g
oats daily, with water ad lib and access to a mineral lick.
Bacteria were initially isolated (in tubes of liquid basic medium) by arbitrarily subculturing 30 to 40 colonies at a time from roll bottles which had contained solidified basic
medium, 2.5 mg L-cysteine/ml,and dilutions of fresh rumen fluid. Gram-stained films from all
cultures were examined, and isolates were roughly sorted and screened for hydrogenation
activity in the basic medium. Dilutions of cultures of active isolates were again cultured in
roll bottles of basic medium. From the highest dilutions, 5 to 10 colonies were subcultured.
Growth in these subcultures was examined and cultures were again tested for hydrogenation
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 03:04:38
I02
P. K E M P , R. W. W H I T E A N D D. J. L A N D E R
activity. This procedure was repeated until the purity of strains associated with hydrogenation activity had been established. The final active isolates were obtained, at levels
of 10' to Io8/ml, from four different animals.
Electron microscopy. Several cultures of each organism were negatively stained with 2 %
sodium phosphotungstate pH 7 - 2and examined in the electron microscope. The cultures
were sampled directly, or after concentrating the bacteria by centrifugation for I min in an
Eppendorf 3200 centrifuge (Eppendorf-Geratebau, Hamburg, Germany).
Fatty acids. I -14C-labelledoleic, linoleic, linolenic and palmitic acids were obtained from
the Radiochemical Centre, Amersham, Buckinghamshire, and stored in benzene solution
(5 pCi/ml) at - 20 "C. A small impurity of trans acids, which increased slowly on storage,
was found in these preparations.
Unlabelled oleic, linoleic, linolenic, cis-vaccenic, palmitic and stearic acids were from
Sigma.
Trans-octadec-9-enoic acid (elaidic acid) was prepared by the selenium isomerization of
oleic acid (Swern & Scanlan, 1953) and purified by thin-layer chromatography of the
methyl ester (see below). A mixture of trans-octadecenoic acids containing mainly transoctadec-I 1-enoic acid was prepared by argentation-t.1.c. of the methyl esters of the products
obtained by incubating organism F2/2 with linoleic acid. Cis-octadec- I 5-enoic acid was
prepared in a similar way from the products obtained by incubating organism ~ 2 / 2with
linolenic acid.
Preparation Of fatty acid substrates. Substrates added to uninoculated media were dried,
from benzene solution under nitrogen, as a film around the lower part of the incubation
tube. Medium was then added and the tube autoclaved, sterile reducing agent added and
the medium equilibrated with CO, and inoculated. When substrate was to be added to a
growing culture or suspension of bacteria it was dissolved in sufficient methanol to give a
solution when sterile buffer (Coleman, 1958) was added. The methanol was removed either
with N, at 40 "C or in a rotary evaporator, and the resulting emulsion finally sonicated for
30 s in a Kerry's ultrasonic cleaning bath type KB80, frequency 80 kHz [Kerry's (Ultrasonics) Ltd, Basildon, Essex].
Extraction of lipids and preparation of fatty acid methyl esters. Incubations were stopped,
and the total lipids extracted, by adding the cultures to 10volumes of chloroform-methanol5 M-HCl (5:6: I, by vol.) containing 50 mg 2,6-di-t-butyl-4-methylphenol/l.
After 20 min,
water (2 vol.) was added and the mixture shaken. The lower, CHC1,-rich phase was taken to
dryness on a rotary evaporator and the lipids hydrolysed by refluxing for I h with 6 %
(w/v) KOH in ethanol-water (19: I, v/v). After cooling, an equal volume of water was
added and the mixture extracted three times with half a volume of light petroleum (b.p.
40 to 60 "C) to remove the unsaponified material. The mixture was then acidified with HCl
and extracted three times with light petroleum. The last three extracts, containing free fatty
acids, were pooled and taken to dryness on a rotary evaporator. The fatty acids were
dissolved in diethyl ether-methanol (9 : I , v/v) and methylated with freshly-distilled diazomethane (Schlenk & Gellerman, 1960).
Gas- and radio-gas-chromatography. Fatty acid methyl esters were analysed using a Pye
104 gas-chromatograph fitted with flame ionization detectors and 5 ft x inch glass
columns packed with either polyethyleneglycol-adipate(PEGA; 10% on Diatomite CAW)
or SE 30 (3 % on Diatomite CQ), all from Pye-Unicam Ltd, Cambridge.
The columns were maintained at either 184 or 197 "C for PEGA and 197 "C for SE30.
Argon-CO, (19: I, v/v) was used as carrier gas at 60 ml/min. A stream splitter allowed 90 to
95 % of the effluent gases to bypass the detector and pass through a heated combustion tube
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 03:04:38
Hydrogenation by rumen bacteria
103
packed with copper oxide followed by iron filings. The resulting 14C02was piped to a Panax
radio-gas counter (Panax Equipment Ltd, Redhill, Surrey).
Position and geometry of double bonds. The fatty acid methyl esters were separated preparatively, according to the degree of unsaturation and the geometry of the double bonds,
by t.1.c. on 250 pm layers of silica gel H (Kieselgel H. Nach Stahl; Merck) containing 14 %
(w/w)AgNO,; benzene-hexane( I : I , v/v) or benzene-diethyl ether (99 : I , v/v) which were used
to develop the plates. Radioactive components were located either with a windowless scanning flow-counter or by autoradiography (EP4 fine grain glass plates; Ilford, Essex). Nonradioactive components on the plates were visualized by viewing under U.V. light after
spraying with 0.2 % (w/v) 2,7-dichlorofluoresceinin ethanol. Appropriate standards chosen
from methyl stearate, oleate, elaidate, linoleate and a-linolenate were spotted on to one
edge of the plate to assist in the identifications. Areas of the plate where fatty acid methyl
esters were detected were carefully scraped off and the esters eluted from the silica gel with
diethyl ether, or, if polar compounds were suspected, with CHC1,-MeOH (I : I, v/v). The
identities of the fatty acid methyl esters in each eluate were checked by g.l.c., and the double
bond positions in the cis and trans monoenoic acid methyl ester bands were determined by
oxidative cleavage at the double bond by the periodate permanganate method of von
Rudloff (I 956). The dicarboxylic acid monomethyl esters produced were methylated and
analysed by g.1.c. on PEGA. A mixture of the dimethyl esters of dicarboxylic acids of
known chain length was used as external standard (dicarboxylic acids C6 to C15; Analabs
Inc., North Haven, Connecticut, U.S.A.). Since the parent C,, acid was usually ~J~C-labelled,
and the dicarboxylic acids were also 1-l4C-labelled, no routine analysis was done to
identify the monocarboxylic acids derived from the methyl end of the chain. Methyl esters of
acids with two or three double bonds were subjected to partial reduction with hydrazine
hydrate in ethanol-H,O (19: I , v/v) (Privett & Nickell, 1966; Kemp & Dawson, 1968;
Roehm & Privett, I 969). The products were separated on AgN0,-impregnated plates using
benzene-hexane (I : I, v/v). The bands containing the cis and trans monoenoic acid methyl
esters were eluted and their bond positions determined as described above. Dienoic acids
which were produced as intermediates during the reduction of trienoic acids were sometimes
isolated and again subjected to partial reduction to yield monoenoic acids. The position and
geometry of the double bonds in these dienoic acids assisted in the determination of the
structure of the parent trienoic acid.
RESULTS
Active bacterial isolates
The Gram-positive coccus ~ 2 / 6and the Gram-positive bacilli F2/2 and w461 could all
hydrogenate linolenic and linoleic acid to a mixture of trans octadecenoic acids but not to
2
~ 3 4 4could hydrogenate oleic 2nd linoleic
stearic acid. The Gram-negative bacilli ~ 2 / and
acid to stearic acid, and linolenic acid to, mainly, cis-octadec-15-enoic acid.
Isolates ~ 2 / 2and ~ 2 / 6were obtained from the same sheep, and isolates ~ 2 / 2 T344
,
and
w461 were each from a different sheep. Originally, w461 and ~ 2 / 2were two-component
cultures, which were taken through at least 20 alternate roll-bottle cultures and subcultures
of individual colonies taken from these, until in each case an active hydrogenating component
and an inactive component were separated. The inactive organisms were discarded since
no enhancement of activity was found in mixed culture. The preliminary description of
w461 (White & Kemp, 1971)was based on this two-component culture before it had been
separated, and was incorrect. The report by Orpin & Munn (1974) from this laboratory,
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 03:04:38
P. K E M P , R . W. W H I T E A N D D. J. L A N D E R
104
Table
I.
Growth and production of organic acids by the bacterial isolates
The basic medium contained glucose, starch and cellobiose. For fermentation tests these were
replaced by the single compounds noted in the first column.
Bacterial isolates
F2/2
w46 I
0'00
0'00
0.03
0'00
0'00
0'00
0-15
0.05
0.04
0'00
0'00
0'00
Substrate
Arabinose
Xylose
Rhamnose
Glucose
Fructose
Mannose
Galactose
Sucrose
Maltose
Lactose
Cellobiose
Melibiose
Trehalose
Raffinose
Starch
Inulin
Dextrin
Glycerol
Erythritol
Adonitol
Mannitol
Dulcitol
Sorbitol
Salicin
Aesculin
Inositol
Lactate
Succinate
Pyruvate
Citrate
Acetate
Fumarate
Pectin
Xylan
Organic acids
produced in the
basic medium
(mmol/IOO ml)
Butyric
Propionic
Acetic
Lactic
Succinic
0'00
O'CO
0'01
0.29
0.46
0.00
0.06
0.0I
0'00
0'00
0.01
0'02
0'00
+, Growth; -, no growth.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 03:04:38
Hydrogenation by rumen bacteria
Fig.
I.
Electron micrographs of (a) ~2/6,
(b) ~ 2 / 2(c)
, ~ 2 / 2and
,
(d)~ 3 4 4 .
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 03:04:38
106
P. KEMP, R. W. WHITE A N D D. J. LANDER
describing bacteriophages W461$1, was also based on the mixed culture and refers to the
inactive component.
Charmterizat ion
The results of growth tests and of organic acid production in the basic medium are set out
in Table I .
None of the five organisms produced H,S. All were catalase-negative, and failed to
reduce gluconate or nitrate, to liquefy gelatin or to digest casein. No decarboxylases to
lysine or ornithine were demonstrated. However, :the test procedure used appeared not
to permit vigorous growth of the organisms. All five organisms were strict anaerobes.
,
I x 0.5 pm (Fig. I a), frequently displayed a
The Gram-positive coccus ~ 2 / 6 measuring
marked streptococcus-like morphology, particularly in a serum medium. It was heat
resistant (60 "C for 30 min) and a narrow capsule could be observed in'indian ink preparations. A grouping precipitation test (Lancefield, 1933) against sera of groups A, B, D, G,
F, H, K, N and P was negative. This amylolytic organism showed some similarities to the
rumen coccus which Provost & Doetsch (1960) designated as a Gram-negative Peptostreptococcus sp. (although their isolate was stated to be heavily granulated in young
cultures). Our strain was morphologically much more variable, and had no proteolytic
activity. Gutierrez et al. (1959) have described rumen organisms similar to Streptococcus
bovis, but heavily capsulated, producing sufficient dextran to form obvious slime in the
presence of specific carbohydrates. When grown in the presence of these carbohydrates
~ 2 / 6failed to produce obvious slime or increased capsules. Strain ~ 2 / 6was designated
Ruminococcus albus, on the basis of morphology in the basic medium, anaerobiosis, and
general fermentative properties.
Strains F2/2 and w461 are both Gram-positive bacilli, 2.5 x 0.5 pm, non-motile, with no
demonstrable flagellum. Strain F2/2 in particular displayed an occasional central swelling
(Fig. I b) which was apparently not a spore. Cultures did not survive heating to 80 "C for
10 min, nor to 60 "C for 30 min. No particular age or condition of culture obviously
affected the numbers of bacilli showing the central swelling. An exactly similar morphology
was described by Walker (1961) for a xylan-utilizing rumen bacillus. Our isolates did not
utilize hay xylan, prepared by D. Jordan in this laboratory by the method of Chanda et al.
(1950). Both strains lost Gram-stain easily. Both were assigned to the genus Eubdcterium
with no species diagnosis.
Isolates ~ 2 / 2(Fig. IC) and T344 (Fig. ~ d were
) Gram-negative flagellated bacilli, 2 to
3 x 0.75 pm. Strain ~ 2 1 2(Fig. I c) was the first strain capable of the biohydrogenation of
linoleic acid and oleic acid to stearic acid to be isolated. Under all conditions of growth,
2
T344 displayed a typical spindle-shaped morphology under the light microboth ~ 2 / and
scope. Vigorous motility was readily observed in young cultures in sealed, hanging-drop
preparations or in the Helba counting-chamber preparations made for total counts of
cultures during the biohydrogenation time-course experiments described later. The method
of using soft agar basic medium in U-tubes inoculated at one end, and observing migration
as a test of motility, did not work for these or any other rumen anaerobes in our hands. A
single, long, sub-polar flagellum was seen in electron microscope preparations (Fig. I c, d ) .
Butyric acid but no acetic or propionic acid was produced from glucose (Table I). On the
basis of the larger size and different morphology, we have separated our isolates quite
clearly from the Butyrivibrio spp. and assigned them to Fusobacteria. Holdeman & Moore
(1973) do not include organisms having a single sub-polar flagellum in their Fusobacterium
genus. All their motile species of this genus produce succinic acid from glucose and they do
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 03:04:38
Hydrogenation by rumen bacteria
70
-
60
-
50
-
40
-
30
-
Vl
T
@
E
C
I=
zx
0.5
1.0
1.5
L-Cysteine (mglml)
Fig.
2.0
2.5
0
5
10
15
20
25
30
Time after inoculation (h)
48
Fig. 3
2
Fig. 2 . Colony counts in (0)
roll bottles and (0)counting flasks (White, 1970), using the basic
medium inoculated with diluted rumen fluid, and varying the levels of L-cysteine. Vertical bars
indicate standard deviations.
Fig. 3. Time course of growth and hydrogenation of oleic acid by ~ 2 / 2 Tubes
.
of the basic
medium containing oleic acid were inoculated with ~ 2 / and
2
incubated at 39 "C. At each interval
two tubes were pooled for hydrogenation tests ( 0 )(details in text) and dilutions from a single tube
were inoculated into roll bottles to assess numbers of viable bacteria (0).The tube contained IO*
bacterialml when inoculated, and 4 x IO* viable bacterialml (=IOO
at 20 h.
x)
not mention Fusucillus. Bryant (1959) refers briefly to Fusobacteria as possible rumen
organisms; Hungate (1966) does not mention this group. Lev (1958) isolated from the rumen
of a cow a Fusfurmis nigrescens which had a particular requirement for vitamin K. Our
isolates showed no stimulation response to menadione, haemin or blood. The two isolates
are assigned to the Fusocillus sp. of Sebald (1962), and may be similar to the Fusocillus sp.
obtained by Barnes & Burton (1970) from the caecum of hibernating ground squirrels.
Strain ~ 3 4 4is a Fusucillus sp. with no species diagnosis. Strain ~ 2 1 2is named as a new
species: Fusucillus babraharnensis nov. spec., Gram-negative, non-sporing bacillus ; motile,
single sub-polar flagellum, strictly anaerobic, spindle-shaped, fermentatively active, produces butyric acid but not succinic acid from glucose, hydrogenates oleic and linoleic acids
to stearic acid, and a-linolenic acid to cis-octadec-15-enoicacid; isolated from the rumen of
a sheep; NCIB number 10838.
All five isolates have been freeze-dried and deposited with the National Collection of
Industrial Bacteria (White, MacKenzie & Bousfield, 1974).
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 03:04:38
72
108
P. KEMP, R. W. WHITE A N D D. J. LANDER
Reducing agents
Figure 2 shows the result, as colony counts in roll bottles and in counting flasks (White,
1g70), of using increasing levels of L-cysteine in basic medium without yeast extract which
had been inoculated with dilutions of fresh rumen contents. Levels of L-cysteine up to 2 3
mg/ml have been used frequently throughout; no observable effect on bacterial growth or
biochemical activity was ever detected.
Dithiothreitol (Cleland, 1964), when used together with L-cysteine, was found by Moore
(1968) to have an additive effect in stimulating surface colony growth of a fastidious
CZostridium sp. It is possible that this was an improved reduction effect similar to that found
here when using high levels of L-cysteine.
Using the basic liquid medium and cultures of the five isolates we compared the results of
using L-cysteine (0.25 mglml) with those obtained with dithiothreitol (I -5 mglml), either
alone or in combination with L-cysteine (0.25 mglml), ascorbic acid (10 mglml), reduced
glutathione (I mg/ml) or sodium thioglycolate (I mglml). Hydrogenation levels after
24 h were almost identical with those for L-cysteine alone, and although viable counts were
not done, growth was not visibly affected by any of the treatments.
Hydrogenation studies
Cultures were grown for 20 to 24 h after heavy inoculation (0.05 ml into 5 ml) from 24 h
cultures. Substrate fatty acids were added at 2 to 20 ,ug/ml. With each organism, increasing
the substrate levels of linoleic and linolenic, but not oleic, acids resulted in an increase of
intermediates relative to the final product, and the higher substrate levels were sometimes
used to obtain sufficient of these intermediates for structural studies.
2
follows the
Figure 3 shows that the hydrogenation of oleic to stearic acid by ~ 2 / closely
growth of the organism. With linoleic acid or linolenic acid as the substrate, the first intermediate (a conjugated dienoic acid or a conjugated trienoic acid) was observed 12 to 15 h
after inoculation. The four other organisms were very similar to ~ 2 / in
2 their time courses
of hydrogenation and growth.
Table 2 shows products recovered and identified after incubating the five organisms with
2 ~ 3 4 were
4
the only organisms of
oleic acid, linoleic acid and linolenic acid. Strains ~ 2 / and
the 30 hydrogenating strains which showed any ability to hydrogenate oleic acid to stearic
acid. There was no evidence to show unequivocally that oleic was hydrogenated via an
intermediate to stearic acid. No cis acids other than the substrate were ever recovered from
the incubations, and oleic acid was apparently hydrogenated more readily to stearic acid
than was cis-octadec-I I -enoic acid, trdns-octadec-9-enoic acid or trans-octadec-I I -enoic
acid.
A polar acid, identified by g.l.c., t.1.c. and mass spectrometry as an hydroxystearic acid,
2
oleic acid but not with other cis or
was always recovered from incubations of ~ 2 / with
trans octadecenoic acids. This polar acid was not metabolized further in our incubation
system and it does not appear that it could be an intermediate. About 5 % of the oleic acid
substrate was recovered as trans-octadec-1I-enoic acid in incubations with T344 but, as
stated above, this acid was not more readily converted to stearic acid than was oleic acid.
The conjugation of linoleic acid to cis,trdns-octadec-g,II-enoic acid and linolenic acid to
cis,trans,cis-octadec-g,I
1,15-enoic acid is the first step in the hydrogenation of these two
acids by all five organisms. Even at very short times of incubation, rarely was more than
10 % of the substrate recovered as the conjugated product. With w461 the yields were so
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 03:04:38
Hydrogenation by rumen bacteria
109
Table 2 . The products and intermediates of hydrogenation
Fatty acids are abbreviated as follows: The number of carbons in the paraffin chain is indicated
before the colon, and the number of double bonds after the colon (e.g. 18:2, for octadecadienoic
acid). This is followed by the position and configuration of the double bonds, numbering from the
carboxyl end of the chain (e.g.r8:3,gc,11t,15~for ~is,t~all~,~is-octadec-~,11,15-enoicacid).
Themajor
end products are indicated in heavy type.
Organism Substrate
Major products
Isomers and other products
P2/2
18 :0 (80 %)
18: 1,9c
(oleic)
18:2,9c, I 2c
(linoleic)
(i) Conjugated I 8 :2,9c, I It
(ii) 18: I , I 1t (70 %)
(iii) 18:o
I ~ : ~ , ~ c , I ~ c ,(i)
I ~Conjugated
c
18:3,9c,11t,1gc
(linolenic)
(ii) 18:2,11t,15c
(iii) 18: 1,15c (85 %)
T344
Oleic
Linoleic
Lino1enic
w46 I
Linoleic
Linolenic
~2/6
(i)
(ii)
(iii)
(i)
(ii)
(iii)
18: 0
18:2,9~,11t
18: 1,rIt (65 %)
18: o
I 8: 3,9c,I I t,15c
18:2,11t,r5c
18 :1,15c (85 yo)
18-OH (20 "/)
9t (15 %),IOt (10 %)
I5t
(10
%),I4t (4 %),I3t
(1
%)
18:1,11t (5 %)
13t,12t,14t,10t (trace 9c)
Trace 9c,15c and I 1c,15c
15t (10 %),14t (5 %)
(i) 18:2,9~,11t
(ii) 18: I (trans 50%) rrt (65 oh) rot (18 %),12t (12 %),gt (2 %)
(cis 50%) IIC (85 %)
IOC (10 %>,9c(1 %)
(i) I 8: 3,9c,1 I t,15c
(by g.l.c.1t.l.c. only)
(ii) I 8 :2, I I t,15c
(iii) 18: I (trans 50%) xrt (65%) lot (20 %),12t (15 %),9t (I %)
(cis 50 Yo) I I C (95 %) I o c (5 %)
Lino1eic
Linolenic
F2/2
Linoleic
Linolenic
(i) 18:2,9~,11t
(ii) 18: I , n t (95 %)
(i) 18:3,gc,r1t,15c
(ii) 18: 2,11t,15c
(iii) 18: 1,rIt (usually poor
conversion - 5 %)
rot (5 %)
Trace ~
C , I ~and
C
others
low that identification was only possible by g.1.c. and t.1.c. No isomeric, conjugated, dienoic
or trienoic acids were detected during the structural analysis of these products.
At the first hydrogenation stage, isomers became apparent when bond positions and
geometry were examined by partial reduction and oxidation. With linoleic acid, by analogy
with results obtained with rumen organisms, the trans-octadec-1I-enoic acid was expected
to be the major octadecenoic acid with a spectrum of close structural isomers in small
amounts.
Table 2 shows that trans-octadec-1I-enoic acid was the predominant isomer found with
four of the organisms. Strain ~ 2 1 6produced a variable mixture of the trans-Io and trans-1 I.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 03:04:38
I I0
P. K E M P , R. W. WHITE A N D D. J. LANDER
The 5 % of cis component found with ~ 2 / 6was richest in the cis-Io acid, the cis-I I isomer
being least abundant. Strain w461 gave an equal distribution between cis and trdnS octadecenoic acids but the cis-11 isomer accounted for 85 % of the cis acids recovered. Good
yields (90 to IOO %) of octadecenoic acids were obtained at 24 h with w461, F 2 / 2 and ~ 2 / 6
from linoleic acid, but ~ 2 / 2and T344, which hydrogenated the bulk of the added linoleic
acid to stearate, rarely yielded more than 15 % of octadecenoic acid.
With all five organisms the first reduction product from linolenic acid was trans,cisoctadec-1 I ,I 5-enoic acid. Small amounts of positional and geometric non-conjugated
dienoic acids were found, the most abundant being cis-9, cis-15 ; cis-10, cis-15 ; cis-I I,
cis-15 ; and trans-10, cis-15. No trans,trans-isomers were found in sufficient quantities to
examine them in detail. Whilst we have not re-incubated these isomers with the organisms,
the amounts isolated appear to be related to the amount of non-conjugated octadecadienoic
acid recovered, suggesting that they were further metabolized.
2
~ 3 4 with
4
linolenic acid
The octadecenoic acids recovered from incubations of ~ 2 / and
were unexpectedly rich in cis-octadec-15-enoic acid (85 %), trans-I 5 being the next predominant isomer. With ~ 2 / 6and w461 the isomers were very similar, both qualitatively and
quantitatively, to those recovered from incubations with linoleic acid. The recoveries of
trans-octadec-1I-enoic acid with F 2 / 2 were very poor. Other isomers may have been present
but these were not recovered in sufficient yield to allow unequivocal characterization.
Re-incubation of products
Non-conjugated dienoic dcid. When re-incubated with the organisms, the non-conjugated
dienoic acid formed by the reduction of linolenic acid by the organisms yielded products
identical to those obtained at 24 h with linolenic acid as the substrate.
Hydroxy-stearic acid. Katz & Keeney (19663) suggested that hydroxy acids were the precursors of the keto-acids found in the rumen of hay-fed steers. The hydroxy-stearic acid
2 the presence of oleic acid was re-incubated with ~ 2 / and
2
with
isolated after growing ~ 2 / in
mixed rumen organisms from sheep. No keto-acid was recovered from ~ 2 / 2incubations,
but there were indications that keto-acid was formed when mixed rumen organisms were
used.
In neither experiment was any stearic acid formed; this would have been expected if an
hydroxy acid was an intermediate in hydrogenation.
Cis-octadec-I 5-enoic acid. On re-incubation, this acid was not converted to either stearic
acid or hydroxy-stearic acid by either mixed rumen organisms from sheep or the isolates
~ 2 / and
2
T344. The substrate was recovered unchanged except for traces of trans-isomer.
Mixed culture hydrogenation
Linolenic acid was incubated for 24 h with a growing, mixed culture of ~ 2 / 2 ~, 2 / 6and
w461. The products were identical with those obtained with ~ 2 / alone
2
(Table 2). From the
numbers of each organism seen in stained preparations, growth of all three bacteria appeared
to be equally good.
Wushed cell suspensions
Strains ~ 2 / 2and ~ 2 / 6were grown for 24 h after inoculating 5 ml of a young culture into
IOO ml of the basic medium without agar. The bacteria were centrifuged using a Sorval
bench centrifuge and washed once with buffer (Coleman, 1958) which had been boiled and
gassed with oxygen-free N2 while cooling. L-Cysteine (0.5 mg/ml) was added. The bacteria
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 03:04:38
Hydrogenation by rumen bacteria
111
were finally suspended in 5 ml of the same buffer and incubated for 4 h at 39 "C with
linolenic acid (2 pg/ml) as substrate. Only very low activity was detected.
Hydrogen donors
An emulsion of linolenic acid (2 pg/ml) was added to a 24 h culture of the organism under
test, together with either sodium formate (45 mM), sodium pyruvate (30 mM) or sodium
succinate (20 mM), and the mixture incubated for 10, 20, 60 and 120 min. In no experiment
was even marginal enhancement of hydrogenation activity found.
DISCUSSION
The numbers in the rumen of each of the five bacteria reported here were between 107
and 108viable bacterialml rumen fluid, indicating that they are true rumen anaerobes and
not chance isolations. Their ability to hydrogenate unsaturated fatty acids may be taken as
further confirmation.
Isolate ~ 2 / has
2 been named as a new species, Fusocillus babrahdmensis. It is possible that
three of the other four are also new species which have not, as far as we are aware, been
isolated elsewhere and have certainly not been associated with hydrogenation activity.
Ruminococcus albus is an established, but somewhat heterologous, rumen species.
From our investigations it is clear that hydrogenation is not an uncommon activity
among rumen bacteria, though in our hands only a few strains of the bacteria isolated were
very active. One can still only speculate on the necessity for hydrogenation in the rumen.
We have not shown that moderate levels of unsaturated fatty acids are inhibitory to, or
necessary for, growth of these bacteria. However, since the isolates were not repeatedly
cultured either in unsaturated fatty acid-free media or in a medium containing high levels
of these acids, the findings are not conclusive. Henderson (1973) demonstrated the variable
effects of adding free fatty acids to pure cultures of six rumen bacteria, and the depression
of methane production and rumen metabolism in vivo by large amounts of unsaturated
fatty acids is well documented (Czerkawski, Blaxter & Wainman, 1966; DeMeyer &
Henderickx, 1967). Whether these effects would be more dramatic in the absence of hydrogenation is not clear. It is possible that some anaerobes can use unsaturated fatty acids as a
hydrogen sink, thus gaining a slight advantage over organisms not able to hydrogenate. Any
rumen organisms to which unsaturated fatty acids are toxic would also benefit.
Qualitatively, the intermediates and products of hydrogenation by the five isolates are
the same as were found in vitro and in vivo with mixed sheep-rumen micro-organisms
(Dawson & Kemp, 1970). Cis-octadec-15-enoic acid, produced in high yields by isolates
~ 2 [ 2and ~ 3 4 4has
, never been observed in large amounts in the rumen. It has been detected
from time to time by t.1.c. and g.l.c., since it has a unique retention time on PEGA at
184 "C (methyl ester 2.23 relative to methyl palmitate).
Trans-octadec-10-enoic acid, produced in large amounts by isolate ~ 2 / 6has
, been found
only in small quantities in the rumen (Katz & Keeney, 1966~).However, on one occasion
this acid was found as the major octadecenoic acid in the depot fat of an apparently normal
five-month-old lamb (W. M. F. Leat and P. Kemp, unpublished observation). Large
amounts of cis octadecenoic acids other than oleic acid have not been found in the rumen.
Thus the finding of equal amounts of cis and trans octadecenoic acids in incubations of
isolate w461 with linoleic and linolenic acid was unexpected. However, since neither the
relative numbers of each hydrogenating organism nor their abilities to hydrogenate in vivo
are known, we cannot estimate the products from a mixed rumen population. Further, one
M I C go
8
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 03:04:38
I12
P. KEMP, R. W. WHITE A N D D. J. L A N D E R
would expect fluctuations in numbers, and a large change in numbers of one or more
hydrogenators due to a dietary or physiological stimulus could account for the unusual
result from one lamb described above. None of our results are incompatible with the
findings with mixed populations from the rumen.
The number of positional and geometric isomers found in the monoenoic and dienoic
acids is greater with mixed populations than with a pure culture. However, we have no
reason to believe that the precursor of the monoenoic acids derived from linolenic is other
than trans,cis-octadec-I I,I yenoic acid. Some of the dienoic acid isomers derive from this
acid, but the monoenoic acids, especially the t~dnSones, seem readily interconverted. The
ability to isomerize these acids would be an advantage, since many isomers occur in plants
and the bacteria might therefore be able to hydrogenate a wider spectrum of substrates. We
have examined only four monoenoic acids, but only the cis-15 isomer appears not to be
further metabolized.
Much work remains to be done to show how the monoenoic acids are hydrogenated to
stearic. It has been suggested (Dawson & Kemp, 1970) that oleic acid, which is hydrogenated
faster than the other monoenoic acids tested, may be more easily converted to a preferred
isomer than either trans-9 or trans-1I - if, indeed, it is not itself the preferred isomer,
Experiments to identify the penultimate product are in progress.
In this connexion the results reported here do not implicate an hydroxy acid as an intermediate, nor suggest why three of the organisms were unable to produce stearic from
linolenic or linoleic acids although producing high yields of its normal precursors, truns
monoenoic acids. It is possible that our growth medium was deficient. In a mixed rumen
population trans monoenoic acids produced by one group of organisms could be converted
to stearic acid by another group. This suggestion has some substance, since the more
unsaturated acids are more rapidly hydrogenated by rumen bacteria to monoenoic acids
than are the monoenoic acids to stearic acid (Dawson & Kemp, 1970). Our experiments
show that isolate PZ/Z has a higher affinity for linolenic acid than ~ 2 1 6and w461 when grown
in mixed culture, since the product with linolenic acid as substrate was mainly the cis-15
monoenoic acid with no stearic acid or trans-1 I monoenoic acid.
It was disappointing to find that the five bacteria were no more resistant to washing than
was Butyrivibriojibrisolvens (Kepler & Tove, 1967)~and that none of the metabolic intermediates which we tested were active as hydrogen donors.
Saturated fatty acids, derived by hydrogenation from dietary unsaturated fatty acids in
the rumen and laid down in ruminant depot fat and secreted in milk, have been implicated
as a factor in human coronary diseases (Jones, 1974). Accordingly, methods have been
developed to decrease the level of saturated acids and increase the unsaturated fatty acids in
ruminant products. Protected seed oils have been fed. These were not hydrolysed in the
rumen, and hence the fatty acids were not hydrogenated but were absorbed in the intestine
and incorporated into carcase and milk fat (Scott & Cook, 1973).
If lipolysis could be prevented or reduced when the animals were fed a normal diet,
hydrogenation would also be reduced. Alternatively, hydrogenation might be controllable if we had a precise knowledge of the metabolic processes leading to the reduction
of free fatty acids by rumen bacteria. At presentwe are trying to identify the hydrogen donors
in hydrogenation, with a view to effecting control at this stage. The effects of such
control measures on the rumen bacterial and protozoal populations will also be carefully
examined to see the overall effects on rumen metabolism.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 03:04:38
Hydrogenation by rumen bacteria
113
We are grateful to Dr R. M. C. Dawson for his interest and encouragement, to Dr
E. A. Munn for preparing electron micrographs, to Mrs M. Black for skilled technical
assistance, and to Mr P. Barker for enzymic estimations of lactic acid.
REFERENCES
BARKER,
J. N. & BRITTON,
H. G. (1957).The enzymatic estimation of L(+)-lactic acid. Journal of Physiology
13% 3p.
BARNES,E. M. & BURTON,
G. C. (1970). The effect of hibernation on the caecal flora of the thirteen-lined
ground squirrel (Citellus tridecemlineatus).Journal of Applied Bacteriology 33, 505-5 14.
BRYANT,
M. P. (1959).Bacterial species of the rumen. Bacteriological Reviews 23, 125-153.
BRYANT,
M.P. & ROBINSON,
I. M. (1961).An improved nonselective culture medium for ruminal bacteria
and its use in determining diurnal variation in numbers of bacteria in the rumen. Journal of Dairy
Science 44,144-1456.
BRYANT,
M. P.& SMALL,
N. (1956).The anaerobic monotrichous butyric acid-producing curved rod-shaped
bacteria of the rumen. Journal of Bacteriology 72, 16-21.
CHANDA,
S. K., HIRST,E. L., JONES, J. K. N. & PERCIVAL,
E. G. V. (1950). The constitution of xylan from
esparto grass (Stipa tenacissima L.). Journal of the Chemical Society, 1289-1297.
CLELAND,
W. W. (1964).Dithiothreitol, a new protective reagent for SH groups. Biochemistry 3, 480-482.
COLEMAN,
G.S.(1958).Maintenance of oligotrich protozoa from the sheep rumen in vitro. Nature, London
182, 1104-1 105.
COWAN,
S. T. & STEEL,
K. J. (1966).Manual for the Identification of Medical Bacteria. Cambridge: Cambridge University Press.
J. W., BLAXTER,
K. L. & WAINMAN,
F. W. (1966). The metabolism of oleic, linoleic and
CZERKAWSKI,
linolenic acids by sheep with reference to their effects on methane production. British Journal of Nutrition 20, 349-362.
DAWSON,
R. M. C., HEMINGTON,
N., GRIME,
D., LANDER,
D. & KEMP,P. (1974).Lipolysisand hydrogenation
of galactolipids and the accumulation of phytanic acid in the rumen. Biochemical Journal 14,169-171.
DAWSON,
R. M.C. & KEMP,P. (1970).Biohydrogenation of dietary fats in ruminants. In Physiology of
Digestion and Metabolism in the Ruminant, pp. 504-518. Edited by A. T.Phillipson. Newcastle: Oriel
Press.
H. K. (1967).The effect of C,, unsaturated fatty acids on methane proDEMEYER,
D. I. & HENDERICKX,
duction in vitro by mixed rumen bacteria. Biochimica et biophysica acta 137,484-497.
GARTON,
G. A., LOUGH,A. K. & VIOQUE,E. (1961).Glyceride hydrolysis and glycerol fermentation by
sheep rumen contents. Journal of General Microbiology 25, 215-225.
GUTIERREZ,
R. E., DAVIS,I. L., LINDAHL,
I. L. & WARWICK;
E. J. (1959).Bacterial changes in the rumen
during the onset of feed-lot bloat of cattle and characteristicsof Peptostreptococcus ekdenii n.sp. Applied
Microbiology 7, 16-22.
HAWKE,
J. C. & SILCOCK,
W. R. (1969).Lipolysis and hydrogenation in the rumen. Biochemical Journal 112,
8
131-132.
HENDERSON,
C . (1973).Effects of fatty acids on pure cultures of rumen bacteria. Journal of Agricultural
Science 81, 107-112.
HOLDEMAN,
L. V. & MOORE,W. E. C. (1973).An Anaerobic Laboratory Manual, 2nd edn. Blacksbury,
Virginia : Virginia Polytechnic Institute.
HUNGATE,
R. W. (1966). The Rumen and its Microbes. New York and London: Academic Press.
JONES,R. J. (1974).Role of dietary fat in health. Journal of the American Oil Chemists Society 51, 251-254.
KATZ,I. & KEENEY,
M. (1966a). Characterization of octadecenoic acids in rumen digesta and rumen
bacteria. Journal of Dairy Science 49,962-966.
KATZ,I. & KEENEY,
M. (19666). Occurrence of ketostearic acids in the rumen. Journal of Dairy Science 49,
967-970.
KEMP,P. & DAWSON,
R. M. C. (1968).Isomerisation of linolenic acid by rumen microorganisms. Biochemical Journal 109,477-478.
KEMP, P. & WHITE,R. W. (1968).The biohydrogenation of linolenic and linoleic acids by bacteria isolated
from an ovine rumen. Biochemical Journal 106,55 P.
KEPLER,C. R., HIRONS,K. P., MCNIELL,J. J. & TOW, S. B. (1966). Intermediates and products of the
biohydrogenation of linoleic acid by Butyrivibrio fibrisolvens. Journal of Biological Chemistry 241,
1350-1 354.
8-2
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 03:04:38
114
P. KEMP, R. W. WHITE A N D D. J. L A N D E R
KEPLER,C. R. & Tom, S. B. (1967).Biohydrogenation of unsaturated fatty acids. 111. Purification and
properties of a linoleateA12-cis,A11-trans-isomerase
from Butyrivibrio jibrisolvens. Journal of Biologicol
Chemistry 242, 5686-5692.
KEPLER,C.R., TUCKER,
W. P. & TOVE,S. B. (1970).Biohydrogenation of unsaturated fatty acids. IV.
Substrate specificity and inhibition of linoleateA12-cis,A11-trans-isomerase
from Butyrivibrio jibrisolvens.
Journal of Biological Chemistry 245, 36I 2-3620.
LANCEFIELD,
R. C. (1933).A serological differentiation of human and other groups of hemolytic streptococci.
Journal of Experimental Medicine 57, 571-595.
LEV,M. (1958).Apparent requirement for vitamin K of rumen strains of Fusiformis nigrescens. Nature,
London 181,203-204.
MILLS,S. C., SCOTT,T. W., RUSSELL,
G. R. & SMITH,R. M. (1970).Hydrogenation of C,, unsaturated
fatty acids by pure cultures of a rumen micrococcus. Australian Journal of Biological Sciences 23,
I 10g-II 13.
MOORE,
W. B. (1968).Solidified media suitable for the cultivation of Clostridium novyi type B. Journal of
General Microbiology 53, 415-423.
NOBLE,
R. C.,STEELE,
W. & MOORE,
J. H. (1969).The incorporation of linoleic acid into the plasma lipids
of sheep given intraruminal infusion of maize oil or free linoleic acid. British Journal of Nutrition 23,
709-7 14.
ORPIN,C. G. & MU", E. A. (1974).The occurrence of bacteriophages in the rumen and their influence on
rumen bacterial populations. Experientiu 30, IOI8-1020.
POLAN,
C.E.,MCNEILL,
J. J. & TOVE,S. B. (1964). Biohydrogenation of unsaturated fatty acids by rumen
bacteria. Journal of Bacteriology 88, 1056-1064.
PRIVETT,
0.S.& NICKELL,
E. C. (1966).Determination of the specific positions of cis and trans double bonds
in polyenes. Lipids I, 98-103.
PROVOST,
P.J. & DOETSCH,
R. N. (1960).An obligate anaerobic amylolytic coccus from the bovine rumen.
Journal of General Microbiology 22, 259-264.
ROEHM,
J. N. & PRIVETT,
0. S. (1969).Improved method for determination of the position of double bonds
in polyenoic fatty acid esters. Journal of Lipid Research 10,245-246.
ROSENFELD,
I. S. & TOVE,S. B. (1971).
Biohydrogenation of unsaturated fatty acids. VI. Source of hydrogen
and stereospecificity of reduction. Journal of Biological Chemistry 246, 3025-5030.
VON RUDLOFF,
E. (1956).Periodate-permanganateoxidations. V. Oxidation of lipids in media containing
organic solvents. Canadian Journal of Chemistry 34, 1413-1418.
RUSSELL,
G.R. & SMITH,R. M. (1968).Reduction of heliotrine by a rumen micro-organism. Australian
Journal of Biological Sciences 21, 1277-1290.
SACHAN,
D.S. & DAVIS,C.L. (1969).Hydrogenation of linoleic acid by a rumen spirochete. Journal of
Bacteriology 98, 300-301.
SCHLENK,
H.& GELLERMAN,
J. L. (1960).Esterification of fatty acids with diazomethane on a small scale.
Analytical Chemistry 32, 1412-1414.
SCOIT, T.W.& COOK,L. J. (1973).Dietary modifications of ruminant milk and meat fats. In The Coronary
Heart Disease and Dietary Fat Controversy, pp. 48-64. Wellington: Editorial Services.
SEBALD,
M. (1962).Gtude sur les bacteries anaerobies Gram-negative asporulCes. Lava1: Barneoud.
SWERN,
D. & SCANLAN,
J. T.(1953).Elaidic acid. Biochemical Preparations 3, I 18-120.
VIVIANI,R. (1970).Metabolism of long-chain fatty acids in the rumen. In Advances in Lipid Research, vol. 8,
pp. 267-345. Edited by R. Paoletti. New York: Academic Press.
WALKER,
D. J. (1961).Isolation and characterisation of a hemicellulose fermenting bacterium from the
rumen. Australian Journal of Agricultural Research 12, 171-175.
WHITE,R. W.(1970).A closed culture vessel for colony counts of anaerobic bacteria, with a note on isolation from a mixed population. Laboratory Practice 19,180.
WHITE,R.W. & KEMP,P. (1971).Sheep rumen bacterial isolates which biohydrogenate unsaturated dietary
fatty acids. Journal of General Microbiology 68, vi-viii.
R. M. C. (1970). Isolation of a rumen bacterium that hydrogenates
WHITE,R. W., KEMP,P. & DAWSON,
oleic acid as well as linoleic acid and linolenic acid. Biochemical Journal 116,767-768.
WHITE,R. W., MACKENZIE,
H. R. & BOUSFIELD,
I. J. (1974).Successful freeze-dryingand retention of biohydrogenation activity of bacteria isolated from the ovine rumen. Journal of Applied Bacteriology 37,vi.
WILDE,P. F. & DAWSON,
R. M. C. (1966).The biohydrogenationof a-linolenicacid and oleic acid by rumen
micro-organisms.Biochemical Journal 98,469-475.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 03:04:38